Toxins appear to be responsible in part for the clinical presentation of some diseases due to bacterial infection. Pertussis toxin, a secretory product of Bordetella pertussis involved in the pathogenesis of whooping cough, and cholera toxin (CT), an etiologic agent in cholera toxin, exert their effect on cells through the ADP-ribosylation of guanine nucleotide- binding (G) proteins that are critical for signaling from cell surface receptors to their intracellular targets (e.g., adenylyl cyclase). Cholera toxin consists of one A subunit, which is an ADP- ribosyltransferase, and five B subunits which bind the toxin to its cell surface receptor, ganglioside GM1. A substrate for the A subunit in disease appears to be the alpha subunit of Gs, the stimulatory G protein of the adenylyl cyclase system and a G protein that may regulate ion flux. Cholera toxin also catalyzes the ADP-ribosylation of simple guanidino compounds (e.g., agmatine, arginine), proteins unrelated to Gsalpha, presumably due to the presence of an accessible arginine residue, and its own A subunit in an auto-ADP-ribosylation reaction. All toxin-catalyzed reactions are enhanced by ARF, consistent with its proposed role as an allosteric activator of CTA1. The present study was undertaken to determine, using in vitro mutagenesis, amino acid residues that may participate in recognition of the G protein by cholera toxin. The assumption was made that recognition of Gsalpha should require the participation of more of the protein structure than would ADP- ribosylation of a simple guanidino compound such as agmatine. Based on this model it was hypothesized that mutations of CTA1 might exist which would preferentially affect ADP-ribosylation of Gs rather than ADP- ribosylation of agmatine. Based on structural analogies between cholera toxin and other bacterial toxins such as pertussis toxin, a number of CTA1 mutations were synthesized in E. coli. Mutations were examined for their ability to ADP-ribosylate agmatine, Gsalpha, and transducin (or Gtalpha) as well as to auto-ADP-ribosylate in various phospholipid/detergent mixtures. Several mutations (e.g., at positions 7, 44, and 112) resulted in overall loss of enzymatic activity. Substitution of asparagine for histidine-70, however, diminished ADP- ribosyl-Gsalpha formation more than it did ADP-ribosylagmatine synthesis. These and other data suggest that histidine-70 may be involved in specific recognition of Gsalpha by cholera toxin.